{"id":2830,"date":"2025-05-27T18:00:14","date_gmt":"2025-05-27T10:00:14","guid":{"rendered":"https:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/"},"modified":"2025-05-27T18:00:14","modified_gmt":"2025-05-27T10:00:14","slug":"%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8","status":"publish","type":"post","link":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/","title":{"rendered":"\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u7814\u7a76\u601d\u8def\/\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362SLAM\u4e2d\u7684\u6a21\u5757"},"content":{"rendered":"<div class=\"article_content clearfix\" id=\"article_content\">\n <link href=\"https:\/\/csdnimg.cn\/release\/blogv2\/dist\/mdeditor\/css\/editerView\/kdoc_html_views-1a98987dfd.css\" rel=\"stylesheet\"\/>\n <link href=\"https:\/\/csdnimg.cn\/release\/blogv2\/dist\/mdeditor\/css\/editerView\/ck_htmledit_views-704d5b9767.css\" rel=\"stylesheet\"\/>\n<div class=\"markdown_views prism-atom-one-dark\" id=\"content_views\">\n  <svg style=\"display: none;\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\n   <path d=\"M5,0 0,2.5 5,5z\" id=\"raphael-marker-block\" stroke-linecap=\"round\" style=\"-webkit-tap-highlight-color: rgba(0, 0, 0, 0);\">\n   <\/path>\n  <\/svg><\/p>\n<p>\n   \u6574\u7406\u4e86\u90e8\u5206\u8fd1\u4e24\u5e74\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u4e00\u4e9b\u7814\u7a76\u6210\u679c\uff08\u53c2\u8003\u77e5\u4e4e\u5e16\u5b50<br \/>\n   <a href=\"https:\/\/www.zhihu.com\/question\/66006923\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n    https:\/\/www.zhihu.com\/question\/66006923<br \/>\n   <\/a><br \/>\n   \u548c\u6ce1\u6ce1\u673a\u5668\u4eba\u516c\u4f17\u53f7\uff0c\u9644\u4e0a\u8bba\u6587\u94fe\u63a5\u548c\u5df2\u627e\u5230\u7684\u6e90\u4ee3\u7801\/\u6570\u636e\u96c6\u94fe\u63a5\uff0c\u5927\u591a\u7b80\u5355\u770b\u4e86\u4e00\u4e0b\u6458\u8981\u3002\u4ec5\u4e3a\u81ea\u5df1\u5b66\u4e60\u6240\u7528\uff0c\u786e\u5b9e\u7ffb\u8bd1\u5f97\u5f88\u70c2\u2026\u2026\u2026\u2026\n  <\/p>\n<h3 id=\"1-\u6df1\u5ea6\u5b66\u4e60\u8ddfslam\u7684\u7ed3\u5408\u70b9\">\n   1. \u6df1\u5ea6\u5b66\u4e60\u8ddfSLAM\u7684\u7ed3\u5408\u70b9<br \/>\n  <\/h3>\n<p>\n   \u6df1\u5ea6\u5b66\u4e60\u548cslam\u7684\u7ed3\u5408\u662f\u8fd1\u51e0\u5e74\u6bd4\u8f83\u70ed\u7684\u4e00\u4e2a\u7814\u7a76\u65b9\u5411\uff0c\u5177\u4f53\u7684\u7814\u7a76\u65b9\u5411\uff0c\u6211\u7b80\u5355\u5206\u4e3a\u4e09\u5757\uff0c\u5982\u4e0b\u3002\n  <\/p>\n<hr\/>\n<h4 id=\"11-\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408slam\u7684\u4e09\u4e2a\u65b9\u5411\">\n   1.1 \u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u4e09\u4e2a\u65b9\u5411<br \/>\n  <\/h4>\n<h5 id=\"\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362\u4f20\u7edfslam\u4e2d\u7684\u4e00\u4e2a\u51e0\u4e2a\u6a21\u5757\">\n   \u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362\u4f20\u7edfSLAM\u4e2d\u7684\u4e00\u4e2a\/\u51e0\u4e2a\u6a21\u5757<br \/>\n  <\/h5>\n<ul>\n<li>\n    \u7279\u5f81\u63d0\u53d6\uff0c\u7279\u5f81\u5339\u914d\uff0c\u63d0\u9ad8\u7279\u5f81\u70b9\u7a33\u5b9a\u6027\uff0c\u63d0\u53d6\u70b9\u7ebf\u9762\u7b49\u4e0d\u540c\u5c42\u7ea7\u7684\u7279\u5f81\u70b9\u3002\n   <\/li>\n<li>\n    \u6df1\u5ea6\u4f30\u8ba1\n   <\/li>\n<li>\n    \u4f4d\u59ff\u4f30\u8ba1\n   <\/li>\n<li>\n    \u91cd\u5b9a\u4f4d\n   <\/li>\n<li>\n    \u5176\u4ed6\n   <\/li>\n<\/ul>\n<h5 id=\"\u5728\u4f20\u7edfslam\u4e4b\u4e0a\u52a0\u5165\u8bed\u4e49\u4fe1\u606f\">\n   \u5728\u4f20\u7edfSLAM\u4e4b\u4e0a\u52a0\u5165\u8bed\u4e49\u4fe1\u606f<br \/>\n  <\/h5>\n<ul>\n<li>\n    \u56fe\u50cf\u8bed\u4e49\u5206\u5272\n   <\/li>\n<li>\n    \u8bed\u4e49\u5730\u56fe\u6784\u5efa\n   <\/li>\n<\/ul>\n<h5 id=\"\u7aef\u5230\u7aef\u7684slam\">\n   \u7aef\u5230\u7aef\u7684SLAM<br \/>\n  <\/h5>\n<p>\n   \u5176\u5b9e\u7aef\u5230\u7aef\u5c31\u4e0d\u80fd\u7b97\u662fSLAM\u95ee\u9898\u4e86\u5427\uff0cSLAM\u662f\u540c\u6b65\u5b9a\u4f4d\u4e0e\u5730\u56fe\u6784\u5efa\uff0c\u7aef\u5230\u7aef\u662f\u8f93\u5165image\u8f93\u51faaction\uff0c\u6ca1\u6709\u5b9a\u4f4d\u548c\u5efa\u56fe\u3002<br \/>\n   <br \/>\n   &#8211; \u673a\u5668\u4eba\u81ea\u4e3b\u5bfc\u822a\uff08\u6df1\u5ea6\u5f3a\u5316\u5b66\u4e60\uff09\u7b49\n  <\/p>\n<hr\/>\n<hr\/>\n<h4 id=\"12-\u76f8\u5173\u7684\u90e8\u5206\u8bba\u6587\u6574\u7406\">\n   1.2 \u76f8\u5173\u7684\u90e8\u5206\u8bba\u6587\u6574\u7406<br \/>\n  <\/h4>\n<h5 id=\"121-\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362\u4f20\u7edfslam\u4e2d\u7684\u4e00\u4e2a\u51e0\u4e2a\u6a21\u5757\">\n   1.2.1 \u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362\u4f20\u7edfSLAM\u4e2d\u7684\u4e00\u4e2a\/\u51e0\u4e2a\u6a21\u5757\u3002<br \/>\n  <\/h5>\n<h6 id=\"\u66ff\u6362\u591a\u4e2a\u6a21\u5757\">\n   \u66ff\u6362\u591a\u4e2a\u6a21\u5757<br \/>\n  <\/h6>\n<ul>\n<li>\n    <strong><br \/>\n     <a href=\"https:\/\/arxiv.org\/pdf\/1704.03489.pdf\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n      Tateno K, Tombari F, Laina I, et al. CNN-SLAM: Real-time dense monocular SLAM with learned depth prediction[J]. arXiv preprint arXiv:1704.03489, 2017.<br \/>\n     <\/a><br \/>\n    <\/strong><br \/>\n    <br \/>\n    <em><br \/>\n     * \u5728\u6709\u9884\u6d4b\u6df1\u5ea6\u4e0b\u7684\u5b9e\u65f6\u5355\u76ee\u7a20\u5bc6SLAM*<br \/>\n    <\/em><br \/>\n    <br \/>\n    \uff08\u8f93\u5165\uff1a\u5f69\u8272\u56fe LSD-SLAM NYUDv2\u6570\u636e\u96c6 ICL-NUIM\u6570\u636e\u96c6\uff09<br \/>\n    <br \/>\n    \u6458\u8981:<br \/>\n    <br \/>\n    \u57fa\u4e8e\u4f7f\u7528\u5377\u79ef\u795e\u7ecf\u7f51\u7edcCNN\u8fdb\u884c\u6df1\u5ea6\u9884\u6d4b\u7684\u6700\u65b0\u8fdb\u5c55\uff0c\u672c\u6587\u7814\u7a76\u4e86\u6df1\u5ea6\u795e\u7ecf\u7f51\u7edc\u751f\u6210\u7684\u6df1\u5ea6\u9884\u6d4b\u5730\u56fe\uff0c\u5982\u4f55\u7528\u4e8e\u7cbe\u786e\u800c\u7a20\u5bc6\u7684\u91cd\u5efa\uff0c\u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u79cd\u76f4\u63a5\u6cd5\u5355\u76eeSLAM\u4e2d\u5f97\u5230\u7684\u6df1\u5ea6\u5ea6\u91cf\uff0c\u5982\u4f55\u4e0eCNN\u9884\u6d4b\u5f97\u5230\u7684\u7a20\u5bc6\u6df1\u5ea6\u5730\u56fe\u81ea\u7136\u5730\u878d\u5408\u5728\u4e00\u8d77\u7684\u65b9\u6cd5\u3002\u6211\u4eec\u7684\u878d\u5408\u65b9\u6cd5\u5728\u56fe\u50cf\u5b9a\u4f4d\u8fd9\u4e00\u5355\u76eeSLAM\u65b9\u6cd5\u6548\u679c\u4e0d\u4f73\u7684\u65b9\u9762\u6709\u4f18\u52bf\u3002\u6bd4\u5982\u8bf4\u4f4e\u7eb9\u7406\u533a\u57df\uff0c\u53cd\u4e4b\u4ea6\u7136\u3002\u6211\u4eec\u8bc1\u660e\u4e86\u6df1\u5ea6\u9884\u6d4b\u5728\u4f30\u8ba1\u91cd\u5efa\u7684\u7edd\u5bf9\u5c3a\u5ea6\u4e2d\u5e94\u7528\u53ef\u4ee5\u514b\u670d\u5355\u76eeSLAM\u7684\u4e3b\u8981\u9650\u5236\u3002\u6700\u540e\uff0c\u6211\u4eec\u63d0\u51fa\u4e86\u9ad8\u6548\u878d\u5408\u7a20\u5bc6SLAM\u4e2d\u5355\u5e27\u5f97\u5230\u7684\u8bed\u4e49\u6807\u7b7e\u7684\u65b9\u6cd5 \uff0c\u4ece\u5355\u89c6\u89d2\u4e2d\u5f97\u5230\u4e86\u8bed\u4e49\u8fde\u8d2f\u7684\u573a\u666f\u91cd\u5efa\u3002\u57fa\u4e8e\u4e24\u4e2a\u53c2\u7167\u6570\u636e\u96c6\u7684\u8bc4\u6d4b\u7ed3\u679c\u8868\u660e\u6211\u4eec\u7684\u65b9\u6cd5\u6709\u826f\u597d\u7684\u9c81\u68d2\u6027\u548c\u51c6\u786e\u6027\u3002<br \/>\n    \n   <\/li>\n<\/ul>\n<hr\/>\n<ul>\n<li>\n    <strong><br \/>\n     <a href=\"https:\/\/arxiv.org\/pdf\/1709.06841.pdf\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n      Li R, Wang S, Long Z, et al. UnDeepVO: Monocular Visual Odometry through Unsupervised Deep Learning[J]. arXiv preprint arXiv:1709.06841, 2017.<br \/>\n     <\/a><br \/>\n    <\/strong><br \/>\n    <br \/>\n    <strong><br \/>\n     UnDeepVO:\u4f7f\u7528\u65e0\u76d1\u7763\u6df1\u5ea6\u5b66\u4e60\u7684\u5355\u76ee\u89c6\u89c9\u91cc\u7a0b\u8ba1<br \/>\n    <\/strong><br \/>\n    <br \/>\n    \uff08\u53cc\u76ee\u56fe\u50cf\u8bad\u7ec3\u6570\u636e\u96c6 \u5355\u76ee\u56fe\u50cf\u6d4b\u8bd5 KITTI\u6570\u636e\u96c6\uff09<br \/>\n    <br \/>\n    \u6458\u8981\uff1a<br \/>\n    <br \/>\n    \u6211\u4eec\u5728\u672c\u6587\u4e2d\u63d0\u51fa\u4e86\u4e00\u79cd\u540d\u53ebUnDeepVO\u7684\u65b0\u578b\u7684\u5355\u76ee\u89c6\u89c9\u91cc\u7a0b\u8ba1\u7cfb\u7edf\uff0cUnDeepVO\u53ef\u4ee5\u4f30\u8ba1\u5355\u76ee\u76f8\u673a\u76846\u81ea\u7531\u5ea6\u4f4d\u59ff\u4ee5\u53ca\u4f7f\u7528\u6df1\u5ea6\u795e\u7ecf\u7f51\u7edc\u4f30\u8ba1\u5355\u76ee\u89c6\u89d2\u7684\u6df1\u5ea6\u3002UnDeepVO\u6709\u4e24\u4e2a\u663e\u8457\u7684\u7279\u6027\uff1a\u4e00\u4e2a\u662f\u65e0\u76d1\u7763\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\uff0c\u53e6\u4e00\u4e2a\u662f\u7edd\u5bf9\u5c3a\u5ea6\u56de\u590d\u3002\u7279\u522b\u7684\uff0c\u6211\u4eec\u4f7f\u7528\u4e86\u53cc\u76ee\u7684\u56fe\u50cf\u5bf9\u8bad\u7ec3UnDeepVO\u6765\u6062\u590d\u5c3a\u5ea6\uff0c\u7136\u540e\u4f7f\u7528\u8fde\u7eed\u7684\u5355\u76ee\u56fe\u50cf\u8fdb\u884c\u4e86\u6d4b\u8bd5\u3002\u56e0\u6b64\uff0cUnDeepVO\u662f\u4e00\u4e2a\u5355\u76ee\u7cfb\u7edf\u3002\u8bad\u7ec3\u7f51\u7edc\u7684\u635f\u5931\u51fd\u6570\u662f\u57fa\u4e8e\u65f6\u95f4\u548c\u7a7a\u95f4\u7a20\u5bc6\u4fe1\u606f\u5b9a\u4e49\u7684\u3002\u56fe\u4e00\u662f\u7cfb\u7edf\u7684\u6982\u89c8\u56fe\u3002\u57fa\u4e8eKITTI\u6570\u636e\u96c6\u7684\u5b9e\u9a8c\u8868\u660eUnDeepVO\u5728\u4f4d\u59ff\u4f30\u8ba1\u65b9\u9762\uff0c\u51c6\u786e\u6027\u9ad8\u4e8e\u5176\u4ed6\u7684\u5355\u76eeVO\u65b9\u6cd5\u3002<br \/>\n    \n   <\/li>\n<\/ul>\n<hr\/>\n<hr\/>\n<h6 id=\"\u7279\u5f81\u76f8\u5173\u7279\u5f81\u63d0\u53d6\u5339\u914d\u7b49\">\n   \u7279\u5f81\u76f8\u5173\uff08\u7279\u5f81\u63d0\u53d6\u5339\u914d\u7b49\uff09<br \/>\n  <\/h6>\n<ul>\n<li>\n    <strong><br \/>\n     <a href=\"https:\/\/arxiv.org\/pdf\/1707.07410.pdf\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n      DeTone D, Malisiewicz T, Rabinovich A. Toward Geometric Deep SLAM[J]. arXiv preprint arXiv:1707.07410, 2017.<br \/>\n     <\/a><br \/>\n    <\/strong><br \/>\n    <br \/>\n    <strong><br \/>\n     \u9762\u5411\u51e0\u4f55\u7684\u6df1\u5ea6SLAM<br \/>\n    <\/strong><br \/>\n    <br \/>\n    \uff08\u4e24\u4e2aCNN\uff0c\u89d2\u70b9\u63d0\u53d6\u548c\u5339\u914d \uff0c\u5b9e\u65f6\uff0c\u5355\u6838CPU30FPS\uff09<br \/>\n    <br \/>\n    \u6458\u8981\uff1a<br \/>\n    <br \/>\n    \u6211\u4eec\u5c55\u793a\u4e86\u4e00\u4e2a\u4f7f\u7528\u4e86\u4e24\u4e2a\u6df1\u5ea6\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u7684\u70b9\u8ddf\u8e2a\u7cfb\u7edf\u3002\u7b2c\u4e00\u4e2a\u7f51\u7edc\uff0cMagicPoint\uff0c\u63d0\u53d6\u5355\u5f20\u56fe\u50cf\u7684\u663e\u8457\u60272D\u70b9\u3002\u8fd9\u4e9b\u63d0\u53d6\u51fa\u6765\u7684\u70b9\u53ef\u4ee5\u7528\u4f5cSLAM\uff0c\u56e0\u4e3a\u4ed6\u4eec\u5728\u56fe\u50cf\u4e2d\u76f8\u4e92\u72ec\u7acb\u4e14\u5747\u5300\u5206\u5e03\u3002\u6211\u4eec\u6bd4\u8f83\u4e86\u8fd9\u4e2a\u7f51\u7edc\u548c\u4f20\u7edf\u7684\u70b9\u68c0\u6d4b\u65b9\u6cd5\uff0c\u53d1\u73b0\u4e24\u8005\u5728\u56fe\u50cf\u6709\u566a\u58f0\u5b58\u5728\u662f\u5b58\u5728\u660e\u663e\u7684\u6027\u80fd\u5dee\u5f02\u3002\u5f53\u68c0\u6d4b\u70b9\u662f\u51e0\u4f55\u7a33\u5b9a\u7684\u65f6\u5019\uff0c\u8f6c\u6362\u4f30\u8ba1\u4f1a\u53d8\u5f97\u66f4\u7b80\u5355\uff0c\u6211\u4eec\u8bbe\u8ba1\u4e86\u7b2c\u4e8c\u4e2a\u7f51\u7edc\uff0c\u540d\u4e3aMagicWarp,\u5b83\u5bf9MagicPoint\u7684\u8f93\u51fa\uff0c\u4e00\u7cfb\u5217\u70b9\u56fe\u50cf\u5bf9\u8fdb\u884c\u64cd\u4f5c\uff0c\u7136\u540e\u4f30\u8ba1\u8ddf\u8f93\u5165\u6709\u5173\u7684\u5355\u5e94\u6027\u3002\u8fd9\u79cd\u8f6c\u6362\u5f15\u64ce\u548c\u4f20\u7edf\u65b9\u6cd5\u7684\u4e0d\u540c\u5728\u4e8e\u5b83\u53ea\u662f\u7528\u70b9\u7684\u5b9a\u4f4d\uff0c\u800c\u6ca1\u6709\u4f7f\u7528\u5c40\u90e8\u70b9\u7684\u63cf\u8ff0\u5b50\u3002\u4e24\u4e2a\u7f51\u7edc\u90fd\u4f7f\u7528\u4e86\u7b80\u5355\u7684\u5408\u6210\u6570\u636e\u8fdb\u884c\u8bad\u7ec3\uff0c\u4e0d\u9700\u8981\u5b89\u89c4\u7684\u5916\u90e8\u76f8\u673a\u5efa\u7acbground truth\u548c\u5148\u8fdb\u7684\u56fe\u5f62\u6e32\u67d3\u6d41\u6c34\u7ebf\u3002\u7cfb\u7edf\u901f\u5ea6\u5feb\u4e14\u8f7b\u91cf\u7ea7\uff0c\u53ef\u4ee5\u5728\u5355\u6838CPU\u4e0a\u8fbe\u523030\u5e27\u6bcf\u79d2\u7684\u901f\u5ea6\u3002<br \/>\n    \n   <\/li>\n<\/ul>\n<hr\/>\n<ul>\n<li>\n    <strong><br \/>\n     <a href=\"https:\/\/arxiv.org\/abs\/1510.05970\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n      Lecun Y. Stereo matching by training a convolutional neural network to compare image patches[M]. JMLR.org, 2016.<br \/>\n     <\/a><br \/>\n    <\/strong><br \/>\n    <br \/>\n    <strong><br \/>\n     \u901a\u8fc7\u8bad\u7ec3\u6bd4\u8f83\u56fe\u50cf\u5757\u7684\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u8fdb\u884c\u7acb\u4f53\u5339\u914d<br \/>\n    <\/strong><br \/>\n    <br \/>\n    \uff08\u8f93\u5165\uff1a\u5de6\u53f3\u56fe KITTI\u6570\u636e\u96c6 Middlebury\u6570\u636e\u96c6\uff09<br \/>\n    <br \/>\n    \u6458\u8981\uff1a<br \/>\n    <br \/>\n    \u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u79cd\u4ece\u5df2\u6821\u6b63\u8fc7\u7684\u56fe\u50cf\u5bf9\u4e2d\u63d0\u53d6\u6df1\u5ea6\u4fe1\u606f\u7684\u65b9\u6cd5\u3002\u6211\u4eec\u7684\u65b9\u6cd5\u4fa7\u91cd\u4e8e\u5927\u591a\u6570stereo\u7b97\u6cd5\u7684\u7b2c\u4e00\u6b65\uff1a\u5339\u914d\u5f00\u9500\u8ba1\u7b97\u3002\u6211\u4eec\u901a\u8fc7\u4f7f\u7528\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u4ece\u5c0f\u56fe\u50cf\u5757\u4e2d\u5b66\u4e60\u76f8\u4f3c\u6027\u5ea6\u91cf\u6765\u89e3\u51b3\u8fd9\u4e2a\u95ee\u9898\u3002\u8bad\u7ec3\u91c7\u7528\u6709\u76d1\u7763\u65b9\u5f0f\uff0c\u4f7f\u7528\u76f8\u4f3c\u548c\u4e0d\u76f8\u4f3c\u7684\u6210\u5bf9\u56fe\u50cf\u5757\u6784\u5efa\u4e86\u4e00\u4e2a\u4e8c\u5206\u7c7b\u6570\u636e\u96c6\u3002\u6211\u4eec\u7814\u7a76\u4e86\u7528\u4e8e\u6b64\u9879\u4efb\u52a1\u7684\u4e24\u79cd\u7f51\u7edc\u67b6\u6784\uff1a\u4e00\u4e2a\u9488\u5bf9\u901f\u5ea6\u8fdb\u884c\u8c03\u6574\uff0c\u53e6\u4e00\u4e2a\u9488\u5bf9\u7cbe\u5ea6\u3002\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u7684\u8f93\u51fa\u88ab\u7528\u6765\u521d\u59cb\u5316stereo\u7acb\u4f53\u5339\u914d\u5f00\u9500\u3002\u5728\u8fd9\u4e4b\u540e\uff0c\u8fdb\u884c\u4e00\u7cfb\u5217\u540e\u5904\u7406\u64cd\u4f5c\uff1a\u57fa\u4e8e\u4ea4\u53c9\u7684\u5f00\u9500\u805a\u5408\uff0c\u534a\u5168\u5c40\u5339\u914d\uff0c\u5de6\u53f3\u56fe\u4e00\u81f4\u6027\u68c0\u9a8c\uff0c\u4e9a\u50cf\u7d20\u589e\u5f3a\uff0c\u4e2d\u503c\u6ee4\u6ce2\u548c\u53cc\u8fb9\u6ee4\u6ce2\u3002\u6211\u4eec\u5728KITTI2012\uff0cKITTI2015\u6570\u636e\u96c6\uff0cMiddlebury\u53cc\u76ee\u6570\u636e\u96c6\u4e0a\u8bc4\u6d4b\u4e86\u81ea\u5df1\u7684\u65b9\u6cd5\uff0c\u7ed3\u679c\u663e\u793a\u6211\u4eec\u7684\u65b9\u6cd5\u4f18\u4e8e\u6b64\u4e09\u4e2a\u6570\u636e\u96c6\u4e0a\u7684\u5176\u4ed6\u540c\u7c7b\u65b9\u6cd5\u3002\n   <\/li>\n<\/ul>\n<hr\/>\n<ul>\n<li>\n    <strong><br \/>\n     <a href=\"https:\/\/arxiv.org\/abs\/1603.09114\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n      Kwang Moo Yi, Eduard Trulls, Vincent Lepetit, et al. LIFT: Learned Invariant Feature Transform[J]. 2016:467-483.<br \/>\n     <\/a><br \/>\n    <\/strong><br \/>\n    <br \/>\n    <strong><br \/>\n     LIFT\uff1a\u901a\u8fc7\u5b66\u4e60\u751f\u6210\u7684\u4e0d\u53d8\u7279\u5f81\u53d8\u6362<br \/>\n    <\/strong><br \/>\n    <br \/>\n    \uff08\u6bd4SIFT\u7279\u5f81\u66f4\u52a0\u7a20\u5bc6\uff0c\u5df2\u5f00\u6e90\uff09<br \/>\n    <br \/>\n    \u6458\u8981\uff1a<br \/>\n    <br \/>\n    \u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u79cd\u65b0\u578b\u7684\u6df1\u5ea6\u7f51\u7edc\u67b6\u6784\uff0c\u5b9e\u73b0\u4e86\u5b8c\u6574\u7684\u7279\u5f81\u70b9\u5904\u7406\u6d41\u6c34\u7ebf\uff1a\u68c0\u6d4b\uff0c\u65b9\u5411\u4f30\u8ba1\u548c\u7279\u5f81\u63cf\u8ff0\u3002\u867d\u7136\u4e4b\u524d\u7684\u5de5\u4f5c\u5df2\u7ecf\u5206\u522b\u6210\u529f\u5730\u89e3\u51b3\u4e86\u8fd9\u51e0\u4e2a\u95ee\u9898\uff0c\u4f46\u6211\u4eec\u5c55\u793a\u4e86\u5982\u4f55\u5c06\u8fd9\u4e09\u4e2a\u95ee\u9898\u7ed3\u5408\u8d77\u6765\uff0c\u901a\u77e5\u4fdd\u6301\u7aef\u5230\u7aef\u7684\u53ef\u5fae\u6027\u3002\u6211\u4eec\u8bc1\u660e\u4e86\u6211\u4eec\u7684\u6df1\u5ea6\u6d41\u6c34\u7ebf\u65b9\u6cd5\uff0c\u6027\u80fd\u4f18\u4e8e\u8bb8\u591a\u57fa\u51c6\u6570\u636e\u96c6\u7684state-of-the-art\u7684\u65b9\u6cd5\uff0c\u4e14\u4e0d\u9700\u8981\u518d\u8bad\u7ec3\u3002\n   <\/li>\n<\/ul>\n<hr\/>\n<hr\/>\n<h6 id=\"\u4f4d\u59ff\u4f30\u8ba1\u6df1\u5ea6\u4f30\u8ba1\">\n   \u4f4d\u59ff\u4f30\u8ba1\uff0c\u6df1\u5ea6\u4f30\u8ba1<br \/>\n  <\/h6>\n<ul>\n<li>\n    <strong><br \/>\n     <a href=\"https:\/\/arxiv.org\/pdf\/1701.08376.pdf\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n      Clark R, Wang S, Wen H, et al. VINet: Visual-Inertial Odometry as a Sequence-to-Sequence Learning Problem[C]\/\/AAAI. 2017: 3995-4001.<br \/>\n     <\/a><br \/>\n    <\/strong><br \/>\n    <br \/>\n    <strong><br \/>\n     VINet:\u5c06\u89c6\u89c9-\u60ef\u6027\u91cc\u7a0b\u8ba1\u770b\u505a\u4e00\u4e2a\u5e8f\u5217\u5230\u5e8f\u5217\u7684\u5b66\u4e60\u95ee\u9898<br \/>\n    <\/strong><br \/>\n    \uff08\u2026\u2026\u8fd9\u4e2a\u600e\u4e48\u7ffb\uff09<br \/>\n    <br \/>\n    \uff08\u4f7f\u7528\u4e86\u56fe\u50cf\u548cIMU\u6570\u636e\uff0cCNN\u548cRNN\uff09<br \/>\n    <br \/>\n    \u6458\u8981\uff1a<br \/>\n    <br \/>\n    \u672c\u6587\u4e2d\u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u79cd\u4f7f\u7528\u89c6\u89c9\u548c\u60ef\u6027\u6570\u636e\u505a\u8fd0\u52a8\u4f30\u8ba1\u7684\uff0c\u6d41\u5f62\u4e0a\u7684\uff1f\u5e8f\u5217\u5230\u5e8f\u5217\u7684\u5b66\u4e60\u65b9\u6cd5\u3002\u5728\u4e2d\u95f4\u7279\u5f81\u8868\u793a\u8fd9\u4e00\u7ea7\u522b\u4e0a\u878d\u5408\u6570\u636e\u7684\u89c6\u89c9-\u60ef\u6027\u91cc\u7a0b\u8ba1\u8fdb\u884c\u7aef\u5230\u7aef\u8bad\u7ec3\uff0c\u662f\u6211\u4eec\u5df2\u77e5\u7684\u6700\u597d\u7684\u65b9\u6cd5\uff08\uff1f\uff09\u3002\u6211\u4eec\u7684\u65b9\u6cd5\u76f8\u6bd4\u4f20\u7edf\u65b9\u6cd5\u6709\u5f88\u591a\u4f18\u52bf\u3002\u5177\u4f53\u6765\u8bf4\uff0c\u5b83\u4e0d\u9700\u8981\u76f8\u673a\u548cIMU\u6570\u636e\u4e4b\u95f4\u8fdb\u884c\u5197\u957f\u4e4f\u5473\u7684\u4eba\u5de5\u540c\u6b65\uff0c\u4e5f\u540c\u6837\u4e0d\u9700\u8981IMU\u548c\u76f8\u673a\u6570\u636e\u4e4b\u95f4\u8fdb\u884c\u4eba\u5de5\u6807\u5b9a\u3002\u53e6\u4e00\u4e2a\u4f18\u70b9\u662f\u6211\u4eec\u7684\u6a21\u578b\u53ef\u4ee5\u81ea\u7136\u4e14\u5de7\u5999\u5730\u7ed3\u5408\u7279\u5b9a\u533a\u57df\u7684\u4fe1\u606f\uff0c\u53ef\u4ee5\u663e\u8457\u51cf\u5c11\u6f02\u79fb\u3002\u5728\u6807\u5b9a\u6570\u636e\u51c6\u786e\u7684\u60c5\u51b5\u4e0b\uff0c\u6211\u4eec\u7684\u65b9\u6cd5\u8ddf\u4f20\u7edf\u7684state-of-the-art\u7684\u65b9\u6cd5\u6548\u679c\u65d7\u9f13\u76f8\u5f53\uff0c\u5728\u5b58\u5728\u6807\u5b9a\u548c\u540c\u6b65\u8bef\u5dee\u7684\u60c5\u51b5\u4e0b\uff0c\u6211\u4eec\u7684\u65b9\u6cd5\u53ef\u4ee5\u901a\u8fc7\u8bad\u7ec3\u8fbe\u5230\u6bd4\u4f20\u7edf\u65b9\u6cd5\u66f4\u597d\u7684\u7684\u6548\u679c\u3002<br \/>\n    \n   <\/li>\n<\/ul>\n<hr\/>\n<ul>\n<li>\n    <strong><br \/>\n     <a href=\"https:\/\/arxiv.org\/pdf\/1603.04992.pdf\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n      Garg R, Vijay K B G, Carneiro G, et al. Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescue[J]. 2016:740-756.<br \/>\n     <\/a><br \/>\n    <\/strong><br \/>\n    <br \/>\n    <strong><br \/>\n     \u7528\u4e8e\u5355\u89c6\u89d2\u6df1\u5ea6\u4f30\u8ba1\u7684\u65e0\u76d1\u7763CNN\uff1a\uff1f\uff1f<br \/>\n    <\/strong><br \/>\n    <br \/>\n    \uff08KITTI\u6570\u636e\u96c6 \u65e0\u76d1\u7763\u5b66\u4e60\uff09<br \/>\n    <br \/>\n    \u6458\u8981\uff1a<br \/>\n    <br \/>\n    \u5f53\u524d\u6df1\u5ea6\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u7684\u4e00\u4e2a\u663e\u8457\u7f3a\u70b9\u5c31\u662f\u9700\u8981\u4f7f\u7528\u5927\u91cf\u4eba\u5de5\u6807\u6ce8\u7684\u6570\u636e\u6765\u8fdb\u884c\u8bad\u7ec3\u3002\u672c\u9879\u7814\u7a76\u4e2d\uff0c\u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u79cd\u65e0\u76d1\u7763\u7684\u6846\u67b6\u6765\u4f7f\u7528\u6df1\u5ea6\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u8fdb\u884c\u5355\u89c6\u89d2\u6df1\u5ea6\u9884\u6d4b\uff0c\u4e0d\u9700\u8981\u5148\u884c\u8bad\u7ec3\u548c\u6807\u6ce8\u8fc7\u7684ground-truth\u6df1\u5ea6\u3002\u6211\u4eec\u901a\u8fc7\u4e00\u79cd\u7c7b\u4f3c\u4e8e\u81ea\u7f16\u7801\u7684\u65b9\u5f0f\u8bad\u7ec3\u7f51\u7edc\u3002\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\uff0c\u6211\u4eec\u8ba4\u4e3a\u6709\u7740\u5fae\u5c0f\u4e14\u5df2\u77e5\u7684\u76f8\u673a\u8fd0\u52a8\u7684\u6e90\u56fe\u50cf\u548c\u76ee\u7684\u56fe\u50cf\u662f\u4e00\u4e2astereo\u5bf9\u3002\u6211\u4eec\u8bad\u7ec3\u5377\u79ef\u7f16\u7801\u5668\u6765\u9884\u6d4b\u6e90\u56fe\u50cf\u7684\u6df1\u5ea6\u56fe\u3002\u4e3a\u6b64\uff0c\u6211\u4eec\u663e\u5f0f\u6784\u9020\u4e86\u4e00\u4e2a\u4f7f\u7528\u9884\u6d4b\u6df1\u5ea6\u548c\u5df2\u77e5\u7684\u89c6\u89d2\u95f4\u4f4d\u79fb\u7684\u76ee\u7684\u56fe\u50cf\u7684inverse warp\u53cd\u53d8\u6362\uff1f\uff0c\u7528\u4e8e\u91cd\u5efa\u6e90\u56fe\u50cf\u3002\u91cd\u5efa\u8fc7\u7a0b\u4e2d\u7684\u5149\u6d4b\u8bef\u5dee\u662f\u7f16\u7801\u5668\u7684\u91cd\u5efa\u635f\u5931\u3002\u4ee5\u8fd9\u6837\u7684\u65b9\u6cd5\u83b7\u53d6\u8bad\u7ec3\u6570\u636e\u6bd4\u540c\u7c7b\u7cfb\u7edf\u8981\u7b80\u5355\u5f97\u591a\uff0c\u4e0d\u9700\u8981\u4eba\u5de5\u6807\u6ce8\u548c\u6df1\u5ea6\u4f20\u611f\u5668\u4e0e\u76f8\u673a\u4e4b\u95f4\u7684\u6807\u5b9a\u3002\u5728KITTI\u6570\u636e\u96c6\u4e0a\uff0c\u6267\u884c\u5355\u89c6\u89d2\u6df1\u5ea6\u4f30\u8ba1\u4efb\u52a1\u65f6\uff0c\u6211\u4eec\u7684\u7f51\u7edc\uff0c\u5728\u4fdd\u8bc1\u76f8\u540c\u6027\u80fd\u60c5\u51b5\u4e0b\uff0c\u8bad\u7ec3\u65f6\u95f4\u6bd4\u5176\u4ed6state-of-the-art\u7684\u6709\u76d1\u7763\u65b9\u6cd5\u5c11\u4e00\u534a\u3002\n   <\/li>\n<\/ul>\n<hr\/>\n<ul>\n<li>\n    <strong><br \/>\n     <a href=\"https:\/\/arxiv.org\/pdf\/1704.07325.pdf\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n      Xu J, Ranftl, Ren\u00e9, Koltun V. Accurate Optical Flow via Direct Cost Volume Processing[J]. 2017.<br \/>\n     <\/a><br \/>\n    <\/strong><br \/>\n    <br \/>\n    \u5149\u6d41\u6cd5\u4e0d\u592a\u5173\u6ce8\uff0c\u8fd9\u4e2a\u540d\u5b57\u4e5f\u662f\u7ffb\u8bd1\u4e0d\u51fa\u6765\u2026\u2026\u2026\u2026<br \/>\n    <br \/>\n    \u82f1\u6587\u6458\u8981\uff1a<br \/>\n    <br \/>\n    We present an optical flow estimation approach that operates on the full four-dimensional cost volume. This direct<br \/>\n    <br \/>\n    approach shares the structural benefits of leading stereo matching pipelines, which are known to yield high accuracy. To this day, such approaches have been considered impractical due to the size of the cost volume. We show that the full four-dimensional cost volume can be constructed in a fraction of a second due to its regularity. We then exploit this regularity further by adapting semi-global matching to the four-dimensional setting. This yields a pipeline that achieves significantly higher accuracy than state-of-the-art optical flow methods while being faster than most. Our approach outperforms all published general-purpose optical flow methods on both Sintel and KITTI 2015 benchmarks.<br \/>\n    \n   <\/li>\n<\/ul>\n<hr\/>\n<ul>\n<li>\n    <strong><br \/>\n     <a href=\"https:\/\/arxiv.org\/pdf\/1611.02174.pdf\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n      Liao Y, Huang L, Wang Y, et al. Parse Geometry from a Line: Monocular Depth Estimation with Partial Laser Observation[J]. 2017.<br \/>\n     <\/a><br \/>\n    <\/strong><br \/>\n    <br \/>\n    <strong><br \/>\n     \u4e00\u6761\u7ebf\u4e0a\u7684\u89e3\u6790\u51e0\u4f55\uff1a\u4f7f\u7528\u90e8\u5206\u6fc0\u5149\u89c2\u6d4b\u7684\u5355\u76ee\u6df1\u5ea6\u4f30\u8ba1<br \/>\n    <\/strong><br \/>\n    <br \/>\n    \uff08\u8f93\u5165\uff1a\u5355\u76ee\u56fe\u50cf\u548c2D\u6fc0\u5149\u8ddd\u79bb\u6570\u636e NYUDv2\u6570\u636e\u96c6 KITTI\u6570\u636e\u96c6\uff09<br \/>\n    <br \/>\n    \u6fc0\u5149\u7684\u4e5f\u4e0d\u592a\u5173\u6ce8\u3002<br \/>\n    <br \/>\n    Abstract\u2014 Many standard robotic platforms are equipped with at least a fixed 2D laser range finder and a monocular camera. Although those platforms do not have sensors for 3D depth sensing capability, knowledge of depth is an essential part in many robotics activities. Therefore, recently, there is an increasing interest in depth estimation using monocular images. As this task is inherently ambiguous, the data-driven estimated depth might be unreliable in robotics applications. In this paper, we have attempted to improve the precision of monocular<br \/>\n    <br \/>\n    depth estimation by introducing 2D planar observation from the remaining laser range finder without extra cost. Specifically, we construct a dense reference map from the sparse laser range data, redefining the depth estimation task as estimating the distance between the real and the reference depth. To solve the problem, we construct a novel residual of residual neural network, and tightly combine the classification and regression losses for continuous depth estimation. Experimental results suggest that our method achieves considerable promotion compared to the state-of-the-art methods on both NYUD2 and KITTI, validating the effectiveness of our method on leveraging the additional sensory information. We further demonstrate the potential usage of our method in obstacle avoidance where our methodology provides comprehensive depth information compared to the solution using monocular camera or 2D laser range finder alone\u3002<br \/>\n    \n   <\/li>\n<\/ul>\n<hr\/>\n<ul>\n<li>\n    <strong><br \/>\n     <a href=\"https:\/\/arxiv.org\/pdf\/1704.07813.pdf\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n      Zhou T, Brown M, Snavely N, et al. Unsupervised learning of depth and ego-motion from video[J]. arXiv preprint arXiv:1704.07813, 2017.<br \/>\n     <\/a><br \/>\n    <\/strong><br \/>\n    <br \/>\n    <strong><br \/>\n     \u89c6\u9891\u6df1\u5ea6\u548c\u81ea\u8fd0\u52a8\u7684\u65e0\u76d1\u7763\u5b66\u4e60<br \/>\n    <\/strong><br \/>\n    SFM-learner<br \/>\n    <br \/>\n    \uff08\u8bad\u7ec3\u4f7f\u7528\u672a\u6807\u6ce8\u5355\u76ee\u89c6\u9891\u7247\u6bb5\uff0c\u5df2\u5f00\u6e90\uff09<br \/>\n    <br \/>\n    \u6458\u8981\uff1a\u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u4e2a\u7528\u975e\u7ed3\u6784\u5316\u89c6\u9891\u5e8f\u5217\u8fdb\u884c\u5355\u76ee\u6df1\u5ea6\u548c\u76f8\u673a\u8fd0\u52a8\u4f30\u8ba1\u7684\u65e0\u76d1\u7763\u5b66\u4e60\u7f51\u7edc\u3002\u548c\u6700\u8fd1\u7684\u51e0\u9879\u7814\u7a76\u76f8\u540c\u7684\u662f\uff0c\u6211\u4eec\u4f7f\u7528\u4e86\u7aef\u5230\u7aef\u7684\u65b9\u6cd5\uff0c\u7528\u89c6\u56fe\u5408\u6210\u4f5c\u4e3a\u76d1\u7763\u4fe1\u53f7\uff0c\u4e0d\u540c\u7684\u662f\uff0c\u6211\u4eec\u7684\u65b9\u6cd5\u662f\u5b8c\u5168\u65e0\u76d1\u7763\u7684\uff0c\u53ea\u9700\u8981\u5c11\u91cf\u7684\u5355\u76ee\u89c6\u9891\u5e8f\u5217\u5373\u53ef\u8bad\u7ec3\u3002\u6211\u4eec\u7684\u65b9\u6cd5\u4f7f\u7528\u4e86\u5355\u89c6\u89d2\u6df1\u5ea6\u548c\u591a\u89c6\u89d2\u4f4d\u59ff\u4e24\u4e2a\u7f51\u7edc\uff0c\u4f7f\u7528\u8ba1\u7b97\u51fa\u7684\u6df1\u5ea6\u548c\u4f4d\u59ff\u5c06\u9644\u8fd1\u89c6\u56fe\u53d8\u6362\u4e3a\u76ee\u6807\u89c6\u56fe\u751f\u6210\u635f\u5931\u51fd\u6570\uff08\uff1f\uff09\u3002\u56e0\u6b64\uff0c\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u7f51\u7edc\u901a\u8fc7\u635f\u5931\u51fd\u6570\u8fde\u63a5\u5728\u4e00\u8d77\uff0c\u4f46\u662f\u6d4b\u8bd5\u65f6\uff0c\u4e24\u4e2a\u7f51\u7edc\u53ef\u4ee5\u72ec\u7acb\u7528\u4e8e\u5e94\u7528\u3002KITTI\u6570\u636e\u96c6\u4e0a\u7684\u7ecf\u9a8c\u8bc4\u6d4b\u8bc1\u660e\u6211\u4eec\u7684\u65b9\u6cd5\u6709\u4ee5\u4e0b\u4f18\u70b9\uff1a1\uff09\u4e0e\u4f7f\u7528ground-truth\u4f4d\u59ff\u6216\u6df1\u5ea6\u8fdb\u884c\u8bad\u7ec3\u7684\u6709\u76d1\u7763\u65b9\u6cd5\u76f8\u6bd4\uff0c\u5728\u4f30\u8ba1\u5355\u76ee\u6df1\u5ea6\u662f\u6548\u679c\u76f8\u5f53\u30022\uff09\u4e0e\u6709\u53ef\u6bd4\u8f83\u8f93\u5165\u8bbe\u7f6e\u7684\u73b0\u6709SLAM\u7cfb\u7edf\u76f8\u6bd4\uff0c\u4f4d\u59ff\u4f30\u8ba1\u6027\u80fd\u826f\u597d\u3002<\/p>\n<p>    <a href=\"https:\/\/github.com\/tinghuiz\/SfMLearner\" rel=\"noopener noreferrer\" target=\"_blank\"><br \/>\n     \u6e90\u4ee3\u7801 https:\/\/github.com\/tinghuiz\/SfMLearner<br \/>\n    <\/a>\n   <\/li>\n<\/ul>\n<hr\/>\n<ul>\n<li>\n    <strong><br \/>\n     <a href=\"https:\/\/arxiv.org\/pdf\/1704.07804.pdf\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n      Vijayanarasimhan S, Ricco S, Schmid C, et al. SfM-Net: Learning of Structure and Motion from Video[J]. arXiv preprint arXiv:1704.07804, 2017.<br \/>\n     <\/a><br \/>\n    <\/strong><br \/>\n    <br \/>\n    <strong><br \/>\n     SFM-Net\uff1a\u4ece\u89c6\u9891\u4e2d\u5b66\u4e60\u7ed3\u6784\u4e0e\u8fd0\u52a8<br \/>\n    <\/strong><br \/>\n    <br \/>\n    SfM-Net\u662fSfM-learner\u7684\u5347\u7ea7\u7248<br \/>\n    <br \/>\n    \u6458\u8981\uff1a<br \/>\n    <br \/>\n    \u6211\u4eec\u63d0\u51fa\u4e86SfM-Net\uff0c\u4e00\u4e2ageometry-aware\u51e0\u4f55\u654f\u611f\uff1f\u7684\u795e\u7ecf\u7f51\u7edc\u7528\u4e8e\u89c6\u9891\u4e2d\u7684\u8fd0\u52a8\u4f30\u8ba1\uff0c\u6b64\u7f51\u7edc\u5206\u89e3\u4e86\u57fa\u4e8e\u573a\u666f\u548c\u5bf9\u8c61\u6df1\u5ea6\u7684\u5e27\u95f4\u50cf\u7d20\u8fd0\u52a8\uff0c\u76f8\u673a\u8fd0\u52a8\uff0c3D\u5bf9\u8c61\u65cb\u8f6c\u548c\u5e73\u79fb\u3002\u7ed9\u5b9a\u4e00\u4e2a\u5e27\u7684\u5e8f\u5217\uff0cSfM-Net\u9884\u6d4b\u6df1\u5ea6\uff0c\u5206\u5272\uff0c\u76f8\u673a\u548c\u521a\u4f53\u8fd0\u52a8\uff0c\u7136\u540e\u5c06\u8fd9\u4e9b\u8f6c\u6362\u4e3a\u7a20\u5bc6\u5e27\u95f4\u8fd0\u52a8\u573a\uff08\u5149\u6d41\uff09\uff0c\u53ef\u5fae\u7684\u626d\u66f2\u5e27\u6700\u540e\u505a\u50cf\u7d20\u5339\u914d\u548c\u53cd\u5411\u4f20\u64ad\u3002\u6a21\u578b\u53ef\u4ee5\u901a\u8fc7\u4e0d\u540c\u7a0b\u5ea6\u7684\u76d1\u7763\u65b9\u6cd5\u8fdb\u884c\u8bad\u7ec3\uff1a1\uff09\u81ea\u76d1\u7763\u7684\u6295\u5f71\u5149\u6d4b\u8bef\u5dee\uff08photometric error\uff09\uff08\u5b8c\u5168\u65e0\u76d1\u7763\uff09\u7684\u65b9\u5f0f\uff0c2\uff09\u7528\u81ea\u8fd0\u52a8\uff08\u76f8\u673a\u8fd0\u52a8\uff09\u8fdb\u884c\u6709\u76d1\u7763\u8bad\u7ec3\u7684\u65b9\u5f0f\uff0c3\uff09\u4f7f\u7528\u6df1\u5ea6\uff08\u6bd4\u5982\u8bf4RGBD\u4f20\u611f\u5668\u63d0\u4f9b\u7684\uff09\u8fdb\u884c\u6709\u76d1\u7763\u8bad\u7ec3\u7684\u65b9\u5f0f\u3002SfM-Net\u63d0\u53d6\u4e86\u6709\u610f\u4e49\u7684\u6df1\u5ea6\u4f30\u8ba1\u5e76\u6210\u529f\u5730\u4f30\u8ba1\u4e86\u5e27\u95f4\u7684\u76f8\u673a\u8fd0\u52a8\u548c\u8bc4\u8bae\u3002\u5b83\u8fd8\u80fd\u5728\u6ca1\u6709\u76d1\u7763\u4fe1\u606f\u63d0\u4f9b\u7684\u60c5\u51b5\u4e0b\uff0c\u6210\u529f\u5206\u5272\u51fa\u573a\u666f\u4e2d\u7684\u8fd0\u52a8\u7269\u4f53\u3002<\/p>\n<\/li>\n<\/ul>\n<hr\/>\n<ul>\n<li>\n    <strong><br \/>\n     <a href=\"https:\/\/arxiv.org\/pdf\/1612.02401.pdf\" rel=\"nofollow noopener noreferrer\" target=\"_blank\"><br \/>\n      Benjamin Ummenhofer, Huizhong Zhou, Jonas Uhrig, et al. DeMoN: Depth and Motion Network for Learning Monocular Stereo[C]\/\/ IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, 2017:5622-5631.<br \/>\n     <\/a><br \/>\n    <\/strong><br \/>\n    <br \/>\n    <strong><br \/>\n     DeMoN\uff1a\u5b66\u4e60\u5355\u53cc\u76ee\uff1f\u6df1\u5ea6\u548c\u8fd0\u52a8\u7684\u7f51\u7edc<br \/>\n    <\/strong><br \/>\n    <br \/>\n    \uff08\u5df2\u5f00\u6e90\uff09<br \/>\n    <br \/>\n    \u6458\u8981\uff1a<br \/>\n    <br \/>\n    \u672c\u6587\u4e2d\u6211\u4eec\u5c06\u8fd0\u52a8\u4e2d\u7684\u7ed3\u6784\u516c\u5f0f\u5316\u5e76\u5c06\u5176\u4f5c\u4e3a\u4e00\u4e2a\u5b66\u4e60\u95ee\u9898\u3002\u6211\u4eec\u7aef\u5230\u7aef\u5730\u8bad\u7ec3\u4e86\u4e00\u4e2a\u5377\u79ef\u7f51\u7edc\u7528\u4e8e\u4ece\u8fde\u7eed\u65e0\u7ea6\u675f\u7684\u56fe\u50cf\u5bf9\u4e2d\u8ba1\u7b97\u6df1\u5ea6\u548c\u76f8\u673a\u8fd0\u52a8\u3002\u6574\u4e2a\u67b6\u6784\u7531\u591a\u5c42\u7f16\u89e3\u7801\u7f51\u7edc\u7ec4\u6210\uff0c\u6838\u5fc3\u90e8\u5206\u662f\u4e00\u4e2a\u53ef\u4ee5\u6539\u8fdb\u81ea\u8eab\u9884\u6d4b\u7684\u8fed\u4ee3\u7f51\u7edc\u3002\u8fd9\u4e2a\u7f51\u7edc\u4e0d\u4ec5\u4f30\u8ba1\u6df1\u5ea6\u548c\u8fd0\u52a8\uff0c\u8fd8\u53ef\u4ee5\u4f30\u8ba1\u8868\u9762\u6cd5\u7ebf\uff0c\u56fe\u50cf\u4e4b\u95f4\u7684\u5149\u6d41\u548c\u5339\u914d\u7684\u810f\u65b0\u90fd\u3002\u57fa\u4e8e\u7a7a\u95f4\u76f8\u5bf9\u5dee\u5f02\u7684\u635f\u5931\u51fd\u6570\u662f\u8fd9\u4e2a\u65b9\u6cd5\u4e2d\u81f3\u5173\u91cd\u8981\u7684\u7ec4\u6210\u90e8\u5206\u3002\u76f8\u6bd4\u4e8e\u4f20\u7edf\u7684\u4ece\u8fd0\u52a8\u4e2d\u5f97\u5230\u4e24\u5e27\u7ed3\u6784\u7684\u65b9\u6cd5\uff0c\u6211\u4eec\u7684\u65b9\u6cd5\u66f4\u52a0\u51c6\u786e\u548c\u9c81\u68d2\u3002\u8ddf\u6d41\u884c\u7684\u4ece\u5355\u5f20\u56fe\u50cf\u83b7\u53d6\u6df1\u5ea6\u7684\u7f51\u7edc\u4e0d\u540c\u7684\u662f\uff0cDeMoN\u5b66\u4e60\u4e86\u5339\u914d\u7684\u6982\u5ff5\uff0c\u80fd\u591f\u5bf9\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u770b\u4e0d\u5230\u7684\u7ed3\u6784\u66f4\u597d\u5730\u6cdb\u5316\u3002<\/p>\n<p>    \u4f7f\u7528pose, depth\u4f5c\u4e3a\u76d1\u7763\u4fe1\u606f\uff0c\u6765\u4f30\u8ba1pose\u548cdepth\u3002<br \/>\n    <br \/>\n    <a href=\"https:\/\/github.com\/lmb-freiburg\/demon\" rel=\"noopener noreferrer\" target=\"_blank\"><br \/>\n     \u6e90\u4ee3\u7801 https:\/\/github.com\/lmb-freiburg\/demon<br \/>\n    <\/a>\n   <\/li>\n<\/ul>\n<hr\/>\n<hr\/>\n<h6 id=\"\u91cd\u5b9a\u4f4d\">\n   \u91cd\u5b9a\u4f4d<br \/>\n  <\/h6>\n<p>\n   \u53ef\u80fd\u91cd\u5b9a\u4f4d\u7528\u6df1\u5ea6\u5b66\u4e60\u6bd4\u8f83\u96be\u505a\u5427\uff0c\u6bd5\u7adf\u662f\u4e2a\u504f\u51e0\u4f55\u7684\u95ee\u9898\uff0c\u6682\u65f6\u4e0d\u592a\u5173\u6ce8<br \/>\n   <br \/>\n   &#8211;<br \/>\n   <a href=\"http:\/\/xlhu.cn\/papers\/Wu17-icra.pdf\" rel=\"nofollow\"><br \/>\n    Wu J, Ma L, Hu X. Delving deeper into convolutional neural networks for camera relocalization[C]\/\/ IEEE International Conference on Robotics and Automation. IEEE, 2017.<br \/>\n   <\/a><br \/>\n   <br \/>\n   &#8211;<br \/>\n   <a href=\"https:\/\/arxiv.org\/pdf\/1505.07427.pdf\" rel=\"nofollow\"><br \/>\n    Alex Kendall, Matthew Grimes, Roberto Cipolla. PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization[J]. 2015, 31:2938-2946.<br \/>\n   <\/a><br \/>\n   <br \/>\n   PoseNet:\u7528\u4e8e\u5b9e\u65f6\u516d\u81ea\u7531\u5ea6\u76f8\u673a\u91cd\u5b9a\u4f4d\u7684\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u3002<br \/>\n   <br \/>\n   PoseNet\u662f2015\u5e74\u7684\u7814\u7a76\u6210\u679c\uff0c\u7b97\u662fSLAM\u8ddf\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408\u7684\u6bd4\u8f83\u6709\u5f00\u521b\u6027\u7684\u6210\u679c\u3002<\/p>\n<p>   <a href=\"https:\/\/github.com\/alexgkendall\/caffe-posenet\"><br \/>\n    \u6e90\u4ee3\u7801 https:\/\/github.com\/alexgkendall\/caffe-posenet<br \/>\n   <\/a>\n  <\/p>\n<hr\/>\n<p>\n   \u53e6\u6709\u4e00\u7bc7\u5f88\u6709\u610f\u601d\u7684\u8bba\u6587<br \/>\n   <br \/>\n   &#8211;<br \/>\n   <strong><br \/>\n    <a href=\"https:\/\/arxiv.org\/pdf\/1705.04838.pdf\" rel=\"nofollow\"><br \/>\n     Vo N, Jacobs N, Hays J. Revisiting IM2GPS in the Deep Learning Era[J]. 2017.<br \/>\n    <\/a><br \/>\n   <\/strong><br \/>\n   <br \/>\n   <strong><br \/>\n    \u6df1\u5ea6\u5b66\u4e60\u65f6\u4ee3\u56fe\u50cf-GPS\u7684\u91cd\u5b9a\u4f4d<br \/>\n   <\/strong><br \/>\n   <br \/>\n   \u601d\u8def\u5f88\u6709\u610f\u601d\uff0c\u4f7f\u7528\u4e00\u5f20\u7167\u7247\u5728\u5168\u4e16\u754c\u8303\u56f4\u5185\u8fdb\u884c\u5b9a\u4f4d\u3002<br \/>\n   \n  <\/p>\n<\/p><\/div>\n<link href=\"https:\/\/csdnimg.cn\/release\/blogv2\/dist\/mdeditor\/css\/editerView\/markdown_views-a5d25dd831.css\" rel=\"stylesheet\"\/>\n <link href=\"https:\/\/csdnimg.cn\/release\/blogv2\/dist\/mdeditor\/css\/style-e504d6a974.css\" rel=\"stylesheet\"\/>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>\u6574\u7406\u4e86\u90e8\u5206\u8fd1\u4e24\u5e74\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u4e00\u4e9b\u7814\u7a76\u6210\u679c\uff08\u53c2\u8003\u77e5\u4e4e\u5e16\u5b50 https:\/\/www.zhihu.com\/ [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":215,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[18],"tags":[],"class_list":["post-2830","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-18"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v24.6 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u7814\u7a76\u601d\u8def\/\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362SLAM\u4e2d\u7684\u6a21\u5757 - \u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408slam\u7684\u7814\u7a76\u601d\u8def-\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\/\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u7814\u7a76\u601d\u8def\/\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362SLAM\u4e2d\u7684\u6a21\u5757 - \u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51\" \/>\n<meta property=\"og:description\" content=\"\u6574\u7406\u4e86\u90e8\u5206\u8fd1\u4e24\u5e74\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u4e00\u4e9b\u7814\u7a76\u6210\u679c\uff08\u53c2\u8003\u77e5\u4e4e\u5e16\u5b50 https:\/\/www.zhihu.com\/ [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408slam\u7684\u7814\u7a76\u601d\u8def-\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\/\" \/>\n<meta property=\"og:site_name\" content=\"\u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51\" \/>\n<meta property=\"article:published_time\" content=\"2025-05-27T10:00:14+00:00\" \/>\n<meta property=\"og:image\" content=\"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/\u8d44\u8baf.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"260\" \/>\n\t<meta property=\"og:image:height\" content=\"180\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"admin@wunen\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u4f5c\u8005\" \/>\n\t<meta name=\"twitter:data1\" content=\"admin@wunen\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4\" \/>\n\t<meta name=\"twitter:data2\" content=\"4 \u5206\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#article\",\"isPartOf\":{\"@id\":\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/\"},\"author\":{\"name\":\"admin@wunen\",\"@id\":\"https:\/\/www.wunen.com\/#\/schema\/person\/d5f7a6cf545656a9c90d507e64452db8\"},\"headline\":\"\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u7814\u7a76\u601d\u8def\/\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362SLAM\u4e2d\u7684\u6a21\u5757\",\"datePublished\":\"2025-05-27T10:00:14+00:00\",\"mainEntityOfPage\":{\"@id\":\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/\"},\"wordCount\":754,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/www.wunen.com\/#organization\"},\"image\":{\"@id\":\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#primaryimage\"},\"thumbnailUrl\":\"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/\u8d44\u8baf.jpg\",\"articleSection\":[\"\u5b66\u4e60\u529e\u516c\"],\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/\",\"url\":\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/\",\"name\":\"\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u7814\u7a76\u601d\u8def\/\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362SLAM\u4e2d\u7684\u6a21\u5757 - \u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51\",\"isPartOf\":{\"@id\":\"https:\/\/www.wunen.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#primaryimage\"},\"image\":{\"@id\":\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#primaryimage\"},\"thumbnailUrl\":\"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/\u8d44\u8baf.jpg\",\"datePublished\":\"2025-05-27T10:00:14+00:00\",\"breadcrumb\":{\"@id\":\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#breadcrumb\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#primaryimage\",\"url\":\"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/\u8d44\u8baf.jpg\",\"contentUrl\":\"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/\u8d44\u8baf.jpg\",\"width\":260,\"height\":180},{\"@type\":\"BreadcrumbList\",\"@id\":\"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\/\/www.wunen.com\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u7814\u7a76\u601d\u8def\/\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362SLAM\u4e2d\u7684\u6a21\u5757\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.wunen.com\/#website\",\"url\":\"https:\/\/www.wunen.com\/\",\"name\":\"\u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51\",\"description\":\"\u8f6f\u4ef6\u8d44\u8baf\u6765\u7269\u5ae9\",\"publisher\":{\"@id\":\"https:\/\/www.wunen.com\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.wunen.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.wunen.com\/#organization\",\"name\":\"\u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51\",\"url\":\"https:\/\/www.wunen.com\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.wunen.com\/#\/schema\/logo\/image\/\",\"url\":\"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/cropped-\u7269\u5ae9-1.png\",\"contentUrl\":\"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/cropped-\u7269\u5ae9-1.png\",\"width\":1024,\"height\":1024,\"caption\":\"\u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51\"},\"image\":{\"@id\":\"https:\/\/www.wunen.com\/#\/schema\/logo\/image\/\"}},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.wunen.com\/#\/schema\/person\/d5f7a6cf545656a9c90d507e64452db8\",\"name\":\"admin@wunen\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.wunen.com\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/d90ec1e3faf77c4d4e66e40c29b85ff6401161e0502f401dae2f0e25b38ce25e?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/d90ec1e3faf77c4d4e66e40c29b85ff6401161e0502f401dae2f0e25b38ce25e?s=96&d=mm&r=g\",\"caption\":\"admin@wunen\"},\"sameAs\":[\"http:\/\/www.wunen.com\"],\"url\":\"http:\/\/www.wunen.com\/index.php\/author\/adminwunen\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u7814\u7a76\u601d\u8def\/\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362SLAM\u4e2d\u7684\u6a21\u5757 - \u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408slam\u7684\u7814\u7a76\u601d\u8def-\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\/","og_locale":"zh_CN","og_type":"article","og_title":"\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u7814\u7a76\u601d\u8def\/\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362SLAM\u4e2d\u7684\u6a21\u5757 - \u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51","og_description":"\u6574\u7406\u4e86\u90e8\u5206\u8fd1\u4e24\u5e74\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u4e00\u4e9b\u7814\u7a76\u6210\u679c\uff08\u53c2\u8003\u77e5\u4e4e\u5e16\u5b50 https:\/\/www.zhihu.com\/ [&hellip;]","og_url":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408slam\u7684\u7814\u7a76\u601d\u8def-\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\/","og_site_name":"\u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51","article_published_time":"2025-05-27T10:00:14+00:00","og_image":[{"width":260,"height":180,"url":"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/\u8d44\u8baf.jpg","type":"image\/jpeg"}],"author":"admin@wunen","twitter_card":"summary_large_image","twitter_misc":{"\u4f5c\u8005":"admin@wunen","\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4":"4 \u5206"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#article","isPartOf":{"@id":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/"},"author":{"name":"admin@wunen","@id":"https:\/\/www.wunen.com\/#\/schema\/person\/d5f7a6cf545656a9c90d507e64452db8"},"headline":"\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u7814\u7a76\u601d\u8def\/\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362SLAM\u4e2d\u7684\u6a21\u5757","datePublished":"2025-05-27T10:00:14+00:00","mainEntityOfPage":{"@id":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/"},"wordCount":754,"commentCount":0,"publisher":{"@id":"https:\/\/www.wunen.com\/#organization"},"image":{"@id":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#primaryimage"},"thumbnailUrl":"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/\u8d44\u8baf.jpg","articleSection":["\u5b66\u4e60\u529e\u516c"],"inLanguage":"zh-Hans","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#respond"]}]},{"@type":"WebPage","@id":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/","url":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/","name":"\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u7814\u7a76\u601d\u8def\/\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362SLAM\u4e2d\u7684\u6a21\u5757 - \u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51","isPartOf":{"@id":"https:\/\/www.wunen.com\/#website"},"primaryImageOfPage":{"@id":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#primaryimage"},"image":{"@id":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#primaryimage"},"thumbnailUrl":"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/\u8d44\u8baf.jpg","datePublished":"2025-05-27T10:00:14+00:00","breadcrumb":{"@id":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#breadcrumb"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/"]}]},{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#primaryimage","url":"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/\u8d44\u8baf.jpg","contentUrl":"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/\u8d44\u8baf.jpg","width":260,"height":180},{"@type":"BreadcrumbList","@id":"http:\/\/www.wunen.com\/index.php\/2025\/05\/27\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%bb%93%e5%90%88slam%e7%9a%84%e7%a0%94%e7%a9%b6%e6%80%9d%e8%b7%af-%e6%88%90%e6%9e%9c%e6%95%b4%e7%90%86%e4%b9%8b%ef%bc%88%e4%b8%80%ef%bc%89%e4%bd%bf%e7%94%a8\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/www.wunen.com\/"},{"@type":"ListItem","position":2,"name":"\u6df1\u5ea6\u5b66\u4e60\u7ed3\u5408SLAM\u7684\u7814\u7a76\u601d\u8def\/\u6210\u679c\u6574\u7406\u4e4b\uff08\u4e00\uff09\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u66ff\u6362SLAM\u4e2d\u7684\u6a21\u5757"}]},{"@type":"WebSite","@id":"https:\/\/www.wunen.com\/#website","url":"https:\/\/www.wunen.com\/","name":"\u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51","description":"\u8f6f\u4ef6\u8d44\u8baf\u6765\u7269\u5ae9","publisher":{"@id":"https:\/\/www.wunen.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.wunen.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"zh-Hans"},{"@type":"Organization","@id":"https:\/\/www.wunen.com\/#organization","name":"\u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51","url":"https:\/\/www.wunen.com\/","logo":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.wunen.com\/#\/schema\/logo\/image\/","url":"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/cropped-\u7269\u5ae9-1.png","contentUrl":"http:\/\/www.wunen.com\/wp-content\/uploads\/2025\/03\/cropped-\u7269\u5ae9-1.png","width":1024,"height":1024,"caption":"\u7269\u5ae9\u8f6f\u4ef6\u8d44\u8baf\u7f51"},"image":{"@id":"https:\/\/www.wunen.com\/#\/schema\/logo\/image\/"}},{"@type":"Person","@id":"https:\/\/www.wunen.com\/#\/schema\/person\/d5f7a6cf545656a9c90d507e64452db8","name":"admin@wunen","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.wunen.com\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/d90ec1e3faf77c4d4e66e40c29b85ff6401161e0502f401dae2f0e25b38ce25e?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/d90ec1e3faf77c4d4e66e40c29b85ff6401161e0502f401dae2f0e25b38ce25e?s=96&d=mm&r=g","caption":"admin@wunen"},"sameAs":["http:\/\/www.wunen.com"],"url":"http:\/\/www.wunen.com\/index.php\/author\/adminwunen\/"}]}},"_links":{"self":[{"href":"http:\/\/www.wunen.com\/index.php\/wp-json\/wp\/v2\/posts\/2830","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/www.wunen.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.wunen.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.wunen.com\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/www.wunen.com\/index.php\/wp-json\/wp\/v2\/comments?post=2830"}],"version-history":[{"count":0,"href":"http:\/\/www.wunen.com\/index.php\/wp-json\/wp\/v2\/posts\/2830\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"http:\/\/www.wunen.com\/index.php\/wp-json\/wp\/v2\/media\/215"}],"wp:attachment":[{"href":"http:\/\/www.wunen.com\/index.php\/wp-json\/wp\/v2\/media?parent=2830"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.wunen.com\/index.php\/wp-json\/wp\/v2\/categories?post=2830"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.wunen.com\/index.php\/wp-json\/wp\/v2\/tags?post=2830"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}